From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:41152) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1atbbW-0000iz-Ii for qemu-devel@nongnu.org; Fri, 22 Apr 2016 09:58:43 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1atbbV-0005WN-Mf for qemu-devel@nongnu.org; Fri, 22 Apr 2016 09:58:42 -0400 Date: Fri, 22 Apr 2016 15:58:30 +0200 From: Kevin Wolf Message-ID: <20160422135830.GF4237@noname.redhat.com> References: <1461322564-17440-1-git-send-email-famz@redhat.com> <20160422134917.GA10620@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20160422134917.GA10620@redhat.com> Subject: Re: [Qemu-devel] [PATCH v2 for-2.6 0/5] block: Fix assertion failure at mirror exit List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: "Michael S. Tsirkin" Cc: Fam Zheng , qemu-devel@nongnu.org, Stefan Hajnoczi , Jeff Cody , Max Reitz , Paolo Bonzini , qemu-block@nongnu.org Am 22.04.2016 um 15:49 hat Michael S. Tsirkin geschrieben: > On Fri, Apr 22, 2016 at 06:55:59PM +0800, Fam Zheng wrote: > > v2: Go along the is_external appraoch. [Kevin] > > > > I tested this series survives the "snapshot + commit" loop reproducer with both > > bonnie++ and RHEL installation in the guest. > > > > This supersedes the "virtio: Register host notifier handler as external" patch > > from yesterday. > > > > The bug was initially reported by Matthew Schumacher as LaunchPad Bug 1570134, > > and nicely bisected by Max Reitz. See patch 2 for the analysis. > > I reviewed patches 1-3 now. > I could merge them but I'd rather not merge patches 4-5 myself though. > Is there a reason why patches 4-5 are in the same series? The fix doesn't work without them because the is_external handling in the main loop is incomplete (i.e. even after aio_disable_external() it would still process the AioHandlers sometimes) > If not, pls submit them separately so block maintainers can merge them. > Alternative have one of the block maintainers merge 1-3, > I sent reviewed-by tags in that case. I think it makes more sense to merge them in a single batch. I can either take v3 through my tree as soon as Fam posts it or Peter can apply the series directly. Kevin